From: Frederic Weisbecker <frederic@kernel.org>
To: Joel Fernandes <joel@joelfernandes.org>
Cc: rcu@vger.kernel.org, linux-kernel@vger.kernel.org,
rushikesh.s.kadam@intel.com, urezki@gmail.com,
neeraj.iitr10@gmail.com, paulmck@kernel.org, rostedt@goodmis.org,
vineeth@bitbyteword.org, boqun.feng@gmail.com
Subject: Re: [PATCH v5 04/18] rcu: Fix late wakeup when flush of bypass cblist happens
Date: Sun, 4 Sep 2022 23:13:50 +0200 [thread overview]
Message-ID: <20220904211350.GA151110@lothringen> (raw)
In-Reply-To: <d24157f4-924c-7113-75ac-ed691191584f@joelfernandes.org>
On Fri, Sep 02, 2022 at 07:58:42PM -0400, Joel Fernandes wrote:
>
>
> On 9/2/2022 7:35 AM, Frederic Weisbecker wrote:
> > On Thu, Sep 01, 2022 at 10:17:06PM +0000, Joel Fernandes (Google) wrote:
> >> When the bypass cblist gets too big or its timeout has occurred, it is
> >> flushed into the main cblist. However, the bypass timer is still running
> >> and the behavior is that it would eventually expire and wake the GP
> >> thread.
> >>
> >> Since we are going to use the bypass cblist for lazy CBs, do the wakeup
> >> soon as the flush happens. Otherwise, the lazy-timer will go off much
> >> later and the now-non-lazy cblist CBs can get stranded for the duration
> >> of the timer.
> >>
> >> This is a good thing to do anyway (regardless of this series), since it
> >> makes the behavior consistent with behavior of other code paths where queueing
> >> something into the ->cblist makes the GP kthread in a non-sleeping state
> >> quickly.
> >>
> >> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
> >> ---
> >> kernel/rcu/tree_nocb.h | 8 +++++++-
> >> 1 file changed, 7 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h
> >> index 0a5f0ef41484..31068dd31315 100644
> >> --- a/kernel/rcu/tree_nocb.h
> >> +++ b/kernel/rcu/tree_nocb.h
> >> @@ -447,7 +447,13 @@ static bool rcu_nocb_try_bypass(struct rcu_data *rdp, struct rcu_head *rhp,
> >> rcu_advance_cbs_nowake(rdp->mynode, rdp);
> >> rdp->nocb_gp_adv_time = j;
> >> }
> >> - rcu_nocb_unlock_irqrestore(rdp, flags);
> >> +
> >> + // The flush succeeded and we moved CBs into the ->cblist.
> >> + // However, the bypass timer might still be running. Wakeup the
> >> + // GP thread by calling a helper with was_all_done set so that
> >> + // wake up happens (needed if main CB list was empty before).
> >> + __call_rcu_nocb_wake(rdp, true, flags)
> >> +
> >
> > Ok so there are two different changes here:
> >
> > 1) wake up nocb_gp as we just flushed the bypass list. Indeed if the regular
> > callback list was empty before flushing, we rather want to immediately wake
> > up nocb_gp instead of waiting for the bypass timer to process them.
> >
> > 2) wake up nocb_gp unconditionally (ie: even if the regular queue was not empty
> > before bypass flushing) so that nocb_gp_wait() is forced through another loop
> > starting with cancelling the bypass timer (I suggest you put such explanation
> > in the comment btw because that process may not be obvious for mortals).
> >
> > The change 1) looks like a good idea to me.
> >
> > The change 2) has unclear motivation. It forces nocb_gp_wait() through another
> > costly loop even though the timer might have been cancelled into some near
> > future, eventually avoiding that extra costly loop. Also it abuses the
> > was_alldone stuff and we may get rcu_nocb_wake with incoherent meanings
> > (WakeEmpty/WakeEmptyIsDeferred) when it's actually not empty.
>
> Yes #2 can be optimized as follows I think on top of this patch, good point:
> =============
> diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h
> index ee5924ba2f3b..24aabd723abd 100644
> --- a/kernel/rcu/tree_nocb.h
> +++ b/kernel/rcu/tree_nocb.h
> @@ -514,12 +514,13 @@ static bool rcu_nocb_try_bypass(struct rcu_data *rdp,
> struct rcu_head *rhp,
> ncbs >= qhimark) {
> rcu_nocb_lock(rdp);
>
> + *was_alldone = !rcu_segcblist_pend_cbs(&rdp->cblist);
> +
> rcu_cblist_set_flush(&rdp->nocb_bypass,
> lazy ? BIT(CB_DEBUG_BYPASS_LAZY_FLUSHED) :
> BIT(CB_DEBUG_BYPASS_FLUSHED),
> (j - READ_ONCE(cb_debug_jiffies_first)));
>
> if (!rcu_nocb_flush_bypass(rdp, rhp, j, lazy, false)) {
> - *was_alldone = !rcu_segcblist_pend_cbs(&rdp->cblist);
> if (*was_alldone)
> trace_rcu_nocb_wake(rcu_state.name, rdp->cpu,
> TPS("FirstQ"));
> @@ -537,7 +538,7 @@ static bool rcu_nocb_try_bypass(struct rcu_data *rdp, struct
> rcu_head *rhp,
> // However, the bypass timer might still be running. Wakeup the
> // GP thread by calling a helper with was_all_done set so that
> // wake up happens (needed if main CB list was empty before).
> - __call_rcu_nocb_wake(rdp, true, flags)
> + __call_rcu_nocb_wake(rdp, *was_all_done, flags)
>
> return true; // Callback already enqueued.
>
That looks right!
}
> =============
>
> > So you may need to clarify the purpose. And I would suggest to make two patches
> > here.
> I guess this change only #2 is no longer a concern? And splitting is not needed
> then as it is only #1.
Sounds good!
Thanks!
next prev parent reply other threads:[~2022-09-04 21:13 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-01 22:17 [PATCH v5 00/18] Implement call_rcu_lazy() and miscellaneous fixes Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 01/18] mm/slub: perform free consistency checks before call_rcu Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 02/18] mm/sl[au]b: rearrange struct slab fields to allow larger rcu_head Joel Fernandes (Google)
2022-09-02 9:26 ` Vlastimil Babka
2022-09-02 9:30 ` Vlastimil Babka
2022-09-02 15:09 ` Joel Fernandes
2022-09-03 13:53 ` Paul E. McKenney
2022-09-01 22:17 ` [PATCH v5 03/18] rcu/tree: Use READ_ONCE() for lockless read of rnp->qsmask Joel Fernandes (Google)
2022-09-06 22:26 ` Boqun Feng
2022-09-06 22:31 ` Joel Fernandes
2022-09-01 22:17 ` [PATCH v5 04/18] rcu: Fix late wakeup when flush of bypass cblist happens Joel Fernandes (Google)
2022-09-02 11:35 ` Frederic Weisbecker
2022-09-02 23:58 ` Joel Fernandes
2022-09-03 15:10 ` Paul E. McKenney
2022-09-04 21:13 ` Frederic Weisbecker [this message]
2022-09-03 14:04 ` Paul E. McKenney
2022-09-03 14:05 ` Joel Fernandes
2022-09-06 3:07 ` Joel Fernandes
2022-09-06 9:48 ` Frederic Weisbecker
2022-09-07 2:43 ` Joel Fernandes
2022-09-01 22:17 ` [PATCH v5 05/18] rcu: Move trace_rcu_callback() before bypassing Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 06/18] rcu: Introduce call_rcu_lazy() API implementation Joel Fernandes (Google)
2022-09-02 15:21 ` Frederic Weisbecker
2022-09-02 23:09 ` Joel Fernandes
2022-09-05 12:59 ` Frederic Weisbecker
2022-09-05 20:18 ` Joel Fernandes
2022-09-05 20:32 ` Joel Fernandes
2022-09-06 8:55 ` Paul E. McKenney
2022-09-06 16:16 ` Joel Fernandes
2022-09-06 17:05 ` Paul E. McKenney
2022-09-03 22:00 ` Joel Fernandes
2022-09-04 21:01 ` Frederic Weisbecker
2022-09-05 20:20 ` Joel Fernandes
2022-09-06 3:05 ` Joel Fernandes
2022-09-06 15:17 ` Frederic Weisbecker
2022-09-06 16:15 ` Joel Fernandes
2022-09-06 16:31 ` Joel Fernandes
2022-09-06 16:38 ` Joel Fernandes
2022-09-06 16:43 ` Joel Fernandes
2022-09-06 19:11 ` Frederic Weisbecker
2022-09-07 2:56 ` Joel Fernandes
2022-09-07 9:56 ` Frederic Weisbecker
2022-09-07 10:03 ` Frederic Weisbecker
2022-09-07 14:01 ` Joel Fernandes
2022-09-07 0:06 ` Joel Fernandes
2022-09-07 9:40 ` Frederic Weisbecker
2022-09-07 13:44 ` Joel Fernandes
2022-09-07 15:38 ` Frederic Weisbecker
2022-09-07 15:39 ` Joel Fernandes
2022-09-21 23:52 ` Joel Fernandes
2022-09-06 18:16 ` Uladzislau Rezki
2022-09-06 18:21 ` Joel Fernandes
2022-09-07 8:52 ` Uladzislau Rezki
2022-09-07 15:23 ` Joel Fernandes
2022-09-03 14:03 ` Paul E. McKenney
2022-09-03 14:05 ` Joel Fernandes
2022-09-01 22:17 ` [PATCH v5 07/18] rcu: shrinker for lazy rcu Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 08/18] rcu: Add per-CB tracing for queuing, flush and invocation Joel Fernandes (Google)
2022-09-02 16:48 ` kernel test robot
2022-09-03 12:39 ` Paul E. McKenney
2022-09-03 14:07 ` Joel Fernandes
2022-09-02 19:01 ` kernel test robot
2022-09-01 22:17 ` [PATCH v5 09/18] rcuscale: Add laziness and kfree tests Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 10/18] rcutorture: Add test code for call_rcu_lazy() Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 11/18] fs: Move call_rcu() to call_rcu_lazy() in some paths Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 12/18] cred: Move call_rcu() to call_rcu_lazy() Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 13/18] security: " Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 14/18] net/core: " Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 15/18] kernel: Move various core kernel usages " Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 16/18] lib: Move call_rcu() " Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 17/18] i915: " Joel Fernandes (Google)
2022-09-01 22:17 ` [PATCH v5 18/18] fork: Move thread_stack_free_rcu() " Joel Fernandes (Google)
2022-09-03 15:22 ` [PATCH v5 00/18] Implement call_rcu_lazy() and miscellaneous fixes Paul E. McKenney
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20220904211350.GA151110@lothringen \
--to=frederic@kernel.org \
--cc=boqun.feng@gmail.com \
--cc=joel@joelfernandes.org \
--cc=linux-kernel@vger.kernel.org \
--cc=neeraj.iitr10@gmail.com \
--cc=paulmck@kernel.org \
--cc=rcu@vger.kernel.org \
--cc=rostedt@goodmis.org \
--cc=rushikesh.s.kadam@intel.com \
--cc=urezki@gmail.com \
--cc=vineeth@bitbyteword.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).